Theory of Financial Intermediation PDF

Empirical Banking Theory of Financial Intermediation Stefano Caiazza 2024-2025  Why do financial intermediaries exist? No need to exist (Neoclassical paradigm) Liquidity Provision (Fractional reserve theory: Old View) Credit Provision (...

Empirical Banking Theory of Financial Intermediation Stefano Caiazza 2024-2025  Why do financial intermediaries exist? No need to exist (Neoclassical paradigm) Liquidity Provision (Fractional reserve theory: Old View) Credit Provision (New View) Transaction costs Diamond and Dybvig Asymmetric information  The Neoclassical Model An adequate theoretical explanation is needed as the importance of banking activity increases. At the end of the nineteenth century, when the so-called “neoclassic" model arose – thanks to the contributions of several economists including Leon Walras, Vilfredo Pareto, Lionel Robbins, Enrico Barone, Kurt Wicksell, Irving Fisher, among others – banks curiously seem to be irrelevant. There is no need for financial intermediaries in a world of perfect financial markets. Funds market in the neoclassical model Perfect Financial Markets Securities can be purchased in any denomination Costless transaction Prices reveal information about the quality of the asset  Theory of disequilibrium – Knut Wicksell (1898) Natural interest rate (rn): the rate compatible with a stable price level. Monetary interest rate (rcr): the actual value of interest rate determined by the banking sector, which contributes to providing financial means for investments. Bank credit can be equal to savings or not. If equal: Credit = Savings → rcr = rn If different: Credit > Savings → rcr < rn in this case: Households: 𝐶𝐶 ↑ Firms: I ↑ (cost < return) __ If C ↑, I ↑→ AD ↑→ P ↑ since Q If P↑ → greater demand for monetary means of payment, i.e., more credit demand. If credit demand ↑ →rcr↑→ rcr = rn For Wicksell, the equality between the two rates will not necessarily happen in the short term, but this difference may last long. Even in the presence of an inflationary process, banks could not necessarily raise interest rates.  The Arrow-Debreu Model Under some assumptions, including the market completeness (state-contingent claims), perfect competition (with market clearing condition), and absence of arbitrage (no free-lunch opportunities), if the following criteria hold: A single agent can not influence prices Conditions for borrowing and lending are equal for all agents There are no discriminatory taxes No economies of scale and scope All financial assets are homogeneous, divisible, and tradable There are no information/transaction/insolvency costs All agents have complete information on all factors and events relevant to the (future) value of the traded financial assets …there is no room for bank intermediaries.  Simplified deterministic Arrow-Debreu model The Consumer The consumer chooses its consumption profile (C1 , C2) and the allocation of its savings S between bank deposits Dh and securities (bonds) Bh, in a way that maximizes his utility function u under the budget constraints: Where ω1 denotes her initial endowment of the consumption good, p denotes the price of C2, Πf and Πb represent, respectively, the profits of the firm and the bank (distributed to the consumer-stockholders at t=2), and r and rD are the interest rates paid by bonds and deposits. Because securities and bank deposits are perfect substitutes, the consumer’s program has an interior solution only when these interest rates are equal → r=rd  Simplified deterministic Arrow-Debreu model The Firm The firm chooses its investment level I and its financing (through bank loans Lf and issuance of securities Bf ) in a way that maximizes its profit: Where f denotes the production function of the representative firm, and rL is the interest rate on bank loans. Again, because bank loans and bonds are here perfect substitutes, Πf has an interior solution only when: r=rl  Simplified deterministic Arrow-Debreu model The Bank The bank chooses its supply of loans Lb, its demand for deposits Db, and its issuance of bonds Bb in a way that maximizes its profit:  Simplified deterministic Arrow-Debreu model General Equilibrium General equilibrium is characterized by a vector of interest rates (r, rL, rD) and three vectors of demand and supply levels — (C1, C2, Bh, Dh) for the consumer, (I, Bf, Lf) for the firm, and (Lb, Bb, Db) for the bank— such that: Each agent behaves optimally (The agent solves respectively) The only possible equilibrium is such that all interest rates are equal: r = rd = rl In a competitive equilibrium: Banks make a zero profit The size and composition of banks’ balance sheets do not affect other economic agents  Schumpeter’s contribution He criticized the idea that the nature of bank deposits is similar to that of depositors whose object is a real asset. Banks create money. So the bank can not be considered as an intermediary that transfers to the borrower a good or a quantity of money that the creditor (depositor) refuses to use. It is necessary to reconsider the concept of credit and savings. Loans determine deposits, not vice versa (due to the deposit multiplier).  Fractional reserve theory of banking (Old View) The fractional reserve theory of banking, later on, called OLD VIEW – developed towards the ‘20s of the last century with the contributions of W.F. Crick, F. von Hayek, J.M. Keynes, A. Phillips, and, in more recent times, P. Samuelson among others – focuses on the role of the banking system to create money through the process of multiple deposit expansion. «Commercial banks do have a special ability to expand credit for a reason that is simple but often overlooked….What is truly unique... about commercial banks is…their distinctive role as issuers of means of payment [which] gives commercial banks a peculiar ability to expand credit» (P. Smith, 1966) Consequence: Banks' functions differ from other intermediaries and must be regulated and supervised. (1892/1894: Severe solvency crisis and distress of the “Banca Romana” caused by having financed the real estate expansion and the demands of the political class).  The “New” View approach The bank is not seen in its ability to create money but as an intermediary. Implication: The bank must issue liabilities (deposits) with the characteristics of size, return, risk, and maturity required by households and, at the same time, issue assets (loans) with size, return, risk, and maturity needed by debtors. Size transformation: banks collect several small deposits and issue large loans. Maturity transformation: banks’ assets and liabilities present different maturity. It is the so- called maturity mismatch. Banks convert demand deposits, withdrawable at sight, into loans with longer maturities. Excessive maturity transformation is therefore undesirable from a financial stability perspective. Risk transformation: banks’ assets and liabilities present different risk profiles. Banks diversify their investments, pool risks, and screen and monitor the quality of borrowers.  The “New” View approach The transformation of maturities and risks does not explain why the bank should perform this function better than the individual industrial firm. In particular, it does not explain why the liabilities issued by intermediaries are better than those issued by corporations. (Commercial Papers https://www.federalreserve.gov/releases/CP/volumestats.htm). To answer these questions, the literature has considered two different approaches: Transaction costs Asymmetric information  Transaction Costs Theory Following Benston and Smith (A transaction cost approach to the theory of financial intermediation, Journal of Finance, 31, 1976), banks’ outputs are neither deposits nor loans but services aimed at reducing transaction costs. «The raison d'etre for this industry [banks] is the existence of transaction costs». Transaction costs can be defined as the costs of transferring resources between markets or participants in the same market. They are associated with counterpart search, contract production, company evaluation (screening), and verification of good performances (monitoring). Financial transactions involve non-financial costs for all participants in the market, i.e., depositors, borrowers, and financial intermediaries. The level and distribution of these costs among the participants are affected by changes in technology, consumer preferences, financial regulations, and internal efficiencies of financial institutions. An intermediary is able to: Achieve economies of scale as a consequence of specialization Obtain the necessary borrower’s financial information at a lower cost because the financial institution is expected to exhibit discretion with that information To reduce search costs  Transaction Costs Theory This theory motivates the existence of banks as they can minimize transaction costs because of the economies of scale achieved in producing financial services. The larger size of an intermediary makes it possible to reduce the incidence of fixed costs The large number of creditors and debtors allows banks to offer an implicit form of insurance to depositors and companies The possible existence of such economies was the main driver of bank gigantism in the 20th century. Intermediaries (could) present economies of scope: The cost of production of financial services decreases as the number of financial goods increases. It could be cheaper to produce a wide range of financial products than to produce less. Diversification (holding positive and negative expected return assets. i.e., loans and deposits) Better management of liquidity risk (credits and deposits) The possible existence of such economies was the main driver of the Universal Bank in the 20th century.  Transaction Costs Theory d = deposit interest rate i = lending rate cs = (depositors) search, information costs to select a bank, and to perform account transactions (deposits, withdrawals). Net return Overall cost cb = (borrowers) costs of negotiating, obtaining, and repaying loans cm = (banks) cost of mobilizing deposits, i.e., resources (labor, capital, materials) c1 = (banks) cost of lending, i.e., processing, monitoring, recovery N = net profits (positive or negative) Banks are firms that use real resources inputs to produce financial services given a particular technology.  Transaction Costs Theory In summary, financial intermediaries are firms that use inputs of real resources to produce financial services given a particular technology. Bank intermediaries have risen to reduce transaction costs, mainly by increasing the size (economies of scale) that reduce fixed transaction costs, offering different financial products (economies of scope), and providing asset transformation services. Banks transform deposits of convenient maturity into loans with larger maturity. “Government regulation increases the transaction costs of financial intermediation principally by restricting financial intermediaries from operating as efficiently as they otherwise would. Furthermore, there appear to be few offsetting benefits for the consumer” (Benston and Smith, 1976).  Diamond and Dybvig (DD) model Bank Runs, Deposit Insurance, and Liquidity, JPE 1983 The model developed by Diamond and Dybvig (1983) analyzes the demand for deposit services (the supply of deposits), showing how the deposit contract benefits families. This contract makes it possible to finance production activities while keeping the savings in highly liquid assets. The fundamental hypothesis is that productive investments take time to generate liquidity and profits. Without banks, households must keep much of their savings as liquidity to face adverse shocks without drastically reducing consumption. Households could finance long-term productive investments only through the share of their savings in excess of their reserves. But these reserves are fundamentally unproductive. By creating liquidity, banks' intermediation increases the system’s efficiency enormously, making it possible to use savings for productive purposes fully. However, agents are subject to liquidity risk.....  Hypotheses of the model Three periods (T0, T1, T2). N risk-averse individuals with an initial endowment equal to 1 (f ≡ fraction of 1). Patient subjects and impatient subjects: private information known in T1. Impatient subjects discover at the beginning of period T1 that they want to consume in T1. Patient subjects find that at the beginning of period T1, they want to consume in T2. The publicly known probability that an individual is impatient is q. Therefore, in an economy with N individuals, the impatient subjects will be Nq and the patient subjects (1-q)N. The economy is characterized by illiquid production technology capable of producing only one homogeneous good. The returns offered by this technology are as follows: T0 T1 T2 Patients -f 0 R with R > f Impatients -f f 0  Solutions 1st possibility: Autarky Each subject behaves as if he were alone, i.e., he consumes an amount of goods corresponding to a unit of money in the intermediate period (T1), if he finds himself impatient, or an amount of goods corresponding to the value of R in the final period (T1), if he finds out himself patient. The risk aversion hypothesis has two consequences. (1) The subjects invest only a fraction of their endowment in technology (i.e., they grant credit to companies), f, holding a share of money to face possible liquidity risks. This behavior reduces resources flowing to businesses, resulting in under-investment and lower economic growth. (2) Each subject would achieve a higher degree of utility if it could operate a risk-sharing by entering into an insurance contract. 2nd possibility: Bank intermediaries The bank intermediary can offer this insurance contract: the bank deposit. Let us suppose that this contract establishes that impatient subjects can consume more than f unit of currency (x1) in T1 and that the remaining part of the wealth available at the end of the period (x2) is devolved to the patient subjects, equal to: N(1-q)x2 = R(N - qNx1) Where the right term is what remains after impatient individuals have been paid in period T1.  Solutions N (1 − q ) x2 = R ( N − qNx1 ) By solving this equation for x2: R ( N − qNx1 ) R(1 − qx1 ) x2 = = N (1 − q ) (1 − q ) If x1> 1, we have that (1 − qx1 ) < 1 , and therefore that x2 < R (1 − q ) It can be shown that the optimal illiquidity insurance contract has the following characteristics: x2 > x1 > 1 x2 x1. Case 2): If the patient subject expects the other patient subjects to ask the bank to reimburse the funds and consume in the intermediate period T1, he jumps to the conclusion that, since x1 > 1, the bank will not have any wealth to distribute in the final period and, indeed, not all requests could be satisfied in the intermediate period. In this context, each patient subject achieves a higher level of utility if he behaves as an impatient subject and asks for the reimbursement of the funds deposited in the intermediate period. The private knowledge of the typology of consumers determines that in the non-cooperative game between the N subjects, two Nash equilibria are possible, one of which replicates the first best solution. In contrast, the bank run constitutes the second.  Possibile solutions to bank runs Suspension of convertibility Under the assumption that the bank adopts the sequential service – first come, first served – and with known q, reimbursing the first Nq subjects and refusing payment to the remaining ones, no run occurs because no patient subject will come to withdraw money. This mechanism can be assimilated into the decision to suspend the convertibility of deposits into currency; a practice followed in the past by some countries (United States, Argentina, for example) to cope with widespread bank runs. Since x2>x1 and no bank run is possible with the mechanism described above, the mere threat of suspending convertibility is sufficient to ensure no patient requests for reimbursement in the intermediate period. However, if q is random, this mechanism cannot avoid bank runs. Deposit Insurance Schemes Regardless of the state of nature, patient subjects are guaranteed reimbursement at time T2. DD argue that deposit insurance must be backed by the Government (or Central Bank). Thanks to the power of taxation, the Government guarantees that each depositor will receive a refund equal to x2. If the commitment is deemed credible, no patient will be interested in anticipating the withdrawal. The Central Bank can print currency. However, no Central Bank wants to make a public commitment to repay the deposits since the discretion of the action would be lost. The Collapse of the Northern Rock Bank  Government-backed deposit insurance DD, pag. 416: «So long as the Government can impose some tax to finance the insurance, no matter how distortionary, there will be no runs and the distorting tax need never be imposed….. the credible promise to provide the insurance means that the promise will not need to be fulfilled. This is in contrast to privately provided deposit insurance. Because insurance companies do not have the power of taxation, they must hold reserves to make their promise credible. This illustrates a reason why the Government may have a natural advantage in providing deposit insurance». Public and private Deposit Insurance Schemes Source: World Bank Financial Safety Net  Asymmetric Information In the Neoclassic World, all agents act with complete certainty. The corporation knows the future cash flows and profit level; the consumer, knowing the future, can calculate the best plan of consumption savings. No unforeseen events, such as crises, changes in habits, etc., are assumed. Information is perfectly distributed among all agents. Summarizing, certainty, and complete information play a key role in that model. Asymmetric information is a situation where: a) Agents operate in a World dominated by the uncertainty b) Each agent has its own information set Information is an expensive asset, difficult to obtain, and therefore not evenly distributed among all operators.  The Market for Lemons. Quality Uncertainty and the Market Mechanism G. Akerlof - The Quartley Journal of Economics, 84, 1970 Four different types of cars characterize the car market: New cars Used cars Good quality cars Bad-quality cars (lemons) The individuals in this market buy a new automobile without knowing whether the car will be good or a lemon. But they do know that with probability q, it is a good car, and with probability (1-q), it is a lemon; by assumption, q is the proportion of good cars produced, and (1 - q) is the proportion of lemons.  The Market for Lemons Suppose the offer price of a good quality car is 2,200$ and that of a bad quality car is 1,000$. Suppose the demand price of a good quality car is 2,400$, and 1,200$ is the demand price of a bad quality car. In a World characterized by imperfect information, the demand for cars is: E ( p D , quality) = q × 2, 400$ + (1 − q ) ×1, 200$ If q would be equal to 0.5: E ( p D , quality) = 0.5 × 2, 400$ + 0.5 ×1, 200$ = 1,800$ The reserve price the buyer is willing to pay is lower than the reserve price the seller is willing to accept for a good quality car.  The Market for Lemons Consequence 1: Good quality car sellers leave the market. Result 1: Bad quality cars drive away good quality cars. Consequence 2: If the buyer understands that there are only bad quality cars in the market, he will not buy cars at all. Result 2: Under certain conditions dependent on: i) The value of q and ii) The buyers and sellers reserve prices no transactions will take place in the market, neither for good nor bad cars, and the car market will cease to exit.  The Market for Lemons Another example cited by Akerlof is the health insurance market. Older people (older than 65) find it very difficult to buy health insurance. So why do insurance companies not raise the price with the risk of increased policy use? Because those who would buy it are the ones who will certainly need it, and as the price increases, only the bad customers will remain. Therefore, selling them to those over 65 at any price is not worthwhile.  Asymmetric Information When some of the participants in the market have information that they are precluded to others, we talk about the existence of information asymmetries. Situations of this type are studied through models called Principal-Agent. In these models, the Principal is the uninformed counterpart, and the Agent is the informed one. We distinguish three classes of models. Those with: Hidden action Hidden information Costly state verification  Asymmetric Information In a model with Hidden action, an individual (or group of individuals), said Principal, has an interest in another individual (or group), said Agent, performing a specific action. For example, in listed companies, shareholders can be considered the Principal and management the Agent. Under perfect information, the Principal remunerates the Agent and observes that the action corresponds to what was agreed upon. Asymmetric information can generate a conflict because the Principal cannot evaluate the Agent’s action. The agent may have incentives to behave undesirably (for the Principal), as the Principal can not prove it and can not enforce the contract. In this case, the Agent is incentivized to exploit the situation to his advantage, taking opportunistic behavior. We talk about moral hazard. However, the Principal knows the temptation of the Agent and, at the time of stipulating the contract, can propose specific contractual solutions to limit the adverse incentives. The solution generally adopted involves linking the remuneration to an observable variable that depends on the Agent’s actions. For example, shareholders can remunerate managers based on profits obtained or, better, on the revaluation of shares, as in stock-option systems.  Asymmetric Information In a model with Hidden information, the Principal can observe the Agent's actions but can not know all the features. For example, an insurance company provides risk coverage policies to large classes of individuals. The latter necessarily knows the risks better than the company. The company is not able to discriminate perfectly. In this case, adverse selection problems arise. If the contract DOES allow discrimination between agents with different characteristics, the Principal selects the agents with whom he wishes to have no relations. In the example above, the insurance company does not provide insurance services to those with more significant risks. If the contract DOES NOT allow discrimination between agents with different characteristics but, for example, offers an equal price to everyone, the Principal will demand a very high insurance premium. In this case, all safer agents leave the market, and total risk will increase. To solve this problem, the contract must provide incentives to allow Agent self-selection. The contract must use signal systems that the Agent can have the incentive to transmit. In the above example, the Principal can create a contract with differentiated tariffs based on the number of previous car accidents, or he can only ensure those willing to invest their resources.  Asymmetric Information Agency Costs Term Agency cost refers to costs that arise from the inefficient relationship between the Principal and the Agent because of asymmetric information. Agency costs can be reduced by an appropriate incentive scheme to align the interests of the Principal and Agent. In the relationship between Stockholders and Managers, the most important and common financial incentives are: Profit sharing: The Agent (i.e., the CEO) receives a percentage of the corporation’s profit if the corporation achieves the targets (e.g., stock price, size, …). Stock options: The Agent (i.e., the CEO) receives a certain number of shares (for free or at a predetermined price) that can be sold from a given period on, if the corporation achieves the targets. Performance bonus: The Agent (i.e., the CEO) receives a predetermined additional compensation if the corporation achieves the targets.  Hidden action and hidden information in the banking industry In the case of bank lending, asymmetric information is a crucial source of credit risk, the risk that the borrower will fail to meet its repayment obligations, causing losses to the bank. If a bank cannot recognize the borrower’s risk, it could increase the interest rate on the loan. Because of this, safer borrowers will leave the market (it is too costly concerning expected return): it is the adverse selection phenomenon. Borrowers who decide to remain in the credit market will invest in riskier projects to ensure the desired expected return: the moral hazard phenomenon.  Credit Rationing Credit rationing is when the demand for credit exceeds the supply at the prevailing interest rate. Dynamic (temporary) rationing: delay in the credit price adjustment. Equilibrium rationing: stable situation. Type I rationing: all customers receive credit, but some (or all) receive less than the desired amount at the market interest rate. Type II rationing: some customers receive credit, and others, identical for some basic features (i.e., risk), are denied credit. Equilibrium type II rationing: the loan application of some borrowers is refused, even if the borrower is willing to pay the market price of the loan contract and is similar (ex-ante) in terms of risk to other borrowers that receive the loan. Hypotheses: Risk-neutral agents Projects of equal size Projects require financing equal to 1 Absence of guarantees Standard debt contract No bankruptcy costs __ The bank knows the average risk Θ of the projects but not the individual risk  The Lender Lender’s profit: ρ L = min {(1 + r ) , X } ≡ min { R, X } ρL Where X is the project’s return, r is the interest rate paid to the lender (bank), and R ≡ (1+r) is the principal and interest R The lender's return is a concave function. The lender’s payoff is nothing more than the R X payoff from a short position in a European put option.  The Lender The lender’s payoff (bank) is the difference between a riskless bond’s payoff, i.e., the bond always pays the same amount independently of the project’s return … ρL X … and the (long) put option payoff, i.e., is a derivative security that gives the right (but not the obligation) to sell a certain amount of underlying asset (X) at an agreed price, called Strike Price (R) at a specific date (European option). ρL R X  The Lender The lender’s payoff is the difference between the riskless bond’s payoff and the long put option’s payoff. ρL ρL R R X R X If the borrower's project is successful and its return is very high, the lender at most always receives the (unitary) capital and the agreed interest. If the project is not successful, the lender sells the put and gets the project value that can be zero in the worst-case scenario. What about the risk?  The lender’s risk At the maturity date (T), the price of a long put option is: P = max {R − X,0} Before the maturity, at time t (1 + δ )  Adverse Selection – The Borrower Since expected profit depends on the risk of the project, there is a level of risk Θ‘ such that only firms with Θ > Θ' undertake the investment. E [π B ] − (1 + δ ) What happens to the borrower if the bank increases the interest rate from r1 to r2? r1 r2 If r increases, the expected profit decreases, causing a right shift in the straight line. E [π B (r ) ] − (1 + δ ) = 0 Θ' Θ '' Θ Firms with low-risk projects will leave the market (adverse selection) E [π (r2 ) ] − (1 + δ ) < 0  Adverse Selection - The Lender What happens to the lender if the interest rate increases? A positive effect and a negative one: Positive: All borrowers will pay more Negative: The safest borrower will leave the market (Adverse Selection) The negative effect can more than offset the positive one. As the profit function of the bank is concave, the higher the risk, the lower the expected profit: the bank’s profit is a decreasing function of the riskiness of the loan.  The Lender’s profit There is a critical interest rate, r*, at which firms with low-risk-return projects leave the credit E (π L ) market, and then the bank's profit decreases Low and drastically. high-risk firms apply E [π L (r*) ] E [π L (r2 ) ] The expected profit function E(πL) is non- Only high-risk monotonic in the interest rate. firms apply r* r2 r  Moral Hazard - shift towards the riskier project Project b is riskier than project a E (π B ) The profit on the less risky project, a, decreases more as the interest rate increases since the borrower is more likely to pay the promised rate. The firm undertakes the project with the highest expected profit; for interest rates to the left of r*, the firm undertakes the Project b less risky project, while for interest rates Project a to the right of r*, it undertakes the riskiest project. r* r  Adverse Selection, Moral Hazard and the Lender’s profit function Expected bank’s profit function Expected bank’s profit function with without asymmetric information asymmetric information E (π L ) E (π L ) ^ r r* r  Borrowers and credit rationing E (π L ) Type 2 borrower: fully satisfied Type 2 borrowers: The credit is granted in full. δ Type 3 borrower: marginal borrower Type 3 borrowers: Some members of this group received credit while others, apparently risk-identifying, were denied. Type 1 borrower: no credit Type 1 borrowers: Credit is denied to these borrowers as they are on the list of untrustworthy debtors. r* r  Stiglitz and Weiss – credit market II I III IV  From the original paper (adapted): ^ Because demand for funds depends on r the interest rate charged by banks, while the supply of funds depends on ρ, the profits on loans, we cannot use a conventional demand/supply curve diagram. The demand for loans is a decreasing function of the interest rate charged to borrowers; this relation LD is drawn in the upper right quadrant. The lower right quadrant draws the non- monotonic relation between the interest charged to borrowers and the expected profit to the bank ρ. In the lower left quadrant, we depict the relation between ρ and the supply of loanable funds Ls. ^ ^ In the upper right quadrant, we plot Ls as a function of r, through the impact of r on the profit. A credit rationing equilibrium exists given the relations drawn in Figure; the demand for loanable ^ ^ funds at r exceeds the supply of loanable funds at r ∗ and any individual bank increasing its ∗ ^ interest rate beyond r ∗ would lower its profit. Z measures the excess demand for funds. Notice that there is an interest rate rm at which the demand for loanable funds equals the supply of loanable funds; however, rm is not an equilibrium interest rate. A bank could increase its ^ profits by charging r ∗ rather than rm: at the lower interest rate, it would attract at least all the borrowers it attracted at rm and make larger profits from each loan.  Possible Solutions Rationing occurs iff the adverse selection effect more than offsets the direct impact of an increase in the interest rate. High-quality (and high-value) collateral. However, richer borrowers who are willing to take more risks (and who are presumably more risk lovers) could benefit in this case. Project co-financed by the entrepreneur with equity capital. It is equivalent to saying that banks make smaller loans. In a multiperiod context, borrowers are often in a position to request additional funds from the lender; a refusal to grant them could jeopardize the repayment of the sums already paid. The low-interest rate for firms with long-term relationships. Regulators forced banks to develop and use internal models to capture individual risk (from Basel 2 on).  The Lending Channel The Lending Channel is typically estimated using as a dependent variable the loans granted to the non-financial sector - in the form of change in loans, log of the stock of loans - and as independent variables: One or more Monetary Policy variables - typically the interest rate or its variation; Bank-level controls: balance sheet variables; Control for unconventional Monetary Policies – Total Asset BC / Nominal GDP Macroeconomic conditions The analysis can be deepened by considering different: Specialization and size of banks Type and maturity of loans Periods of time. Before and after crises: financial / sovereign debt / pandemic  Estimation of credit rationing One of the most widespread analyses of Credit Rationing is to analyze the determinants of rationing. The dependent variable is often a 0-1 dummy obtained from company questionnaires regarding the "perception" of being rationed. This variable is regressed on a set of variables which typically include: Size of the company Type of business Reason for requesting financing Guarantees issued Balance Sheet variables Macroeconomic variables (inflation; unemployment; GDP; official rates) The analysis can be deepened by considering different: Specialization and size of banks Type and maturity of loans Periods of time. Before and after crises: financial / sovereign debt / pandemic  Costly State Verification: Delegated Monitoring Diamond (1984) suggested that banks can better deal with the effects of asymmetric information. In the presence of asymmetric information, it is mandatory screening (ex-ante) of the projects – it requires specific knowledge that is difficult to find in an individual investor, also because of the cost of access to the technologies capable of performing it – and, once financed, monitoring (ex-post), that is to monitor the ongoing soundness of the project. Specifically, verification of the state of the project could be expensive and challenging to implement for a single small investor. So, monitoring carried out by an intermediary is socially less costly than the individual investor would endure, given the presence of an illiquid asset (loans) and a liability consisting of standard debt instruments of tiny unit size (deposits). Depositors then delegate to the intermediary the monitoring of projects.  Delegated Monitoring Suppose monitoring is efficient (monitoring cost is lower than auditing cost). In that case, investors are small, and investment is profitable; financial intermediation (delegated monitoring) dominates direct lending as soon as n is large enough (diversification). Monitoring Delegated Monitoring